Molecular-Scale Visualization of Steric Effects of Ligand Binding to Reconstructed Au(111) Surfaces.

Direct imaging of single molecules at nanostructured interfaces is a grand challenge with potential to enable new, precise material architectures and technologies. Of particular interest are the structural morphology and spectroscopic signatures of the adsorbed molecule, where modern probes are only now being developed with the necessary spatial and energetic resolution to provide detailed information at the molecule-surface interface. Here, we directly characterize the adsorption of individual m-terphenyl isocyanide ligands on a reconstructed Au(111) surface through scanning tunneling microscopy and inelastic electron tunneling spectroscopy. The site-dependent steric pressure of the various surface features alters the vibrational fingerprints of the m-terphenyl isocyanides, which are characterized with single-molecule precision through joint experimental and theoretical approaches. This study provides molecular-level insights into the steric-pressure-enabled surface binding selectivity as well as its effect on the chemical properties of individual surface-binding ligands.

VIRMA promotes the progression of head and neck squamous cell carcinoma by regulating UBR5 mRNA and m6A levels.

Head and neck squamous cell carcinoma (HNSCC) is a globally prevalent and lethal cancer form, whose precise mechanisms remain incompletely understood. Increasing evidence suggests that N6-methyladenosine (m6A) plays a crucial role in cancer progression. This study aimed to explore the biological function of m6A modification and vir-like m6A methyltransferase associated (VIRMA) in HNSCC. We conducted an analysis of VIRMA expression in HNSCC cells using The Cancer Genome Atlas (TCGA) database and employed reverse transcription quantitative polymerase chain reaction (RT-qPCR) and western blotting to assess its expression levels in HNSCC cell lines. Additionally, m6A levels in HNSCC cells were quantified, and the correlation between VIRMA expression levels and the clinical and pathological features of other genes was analyzed. Upon knocking down VIRMA levels, we assessed HNSCC cell proliferation, migration, and invasion and validated downstream genes using RT-qPCR and western blot. Our findings suggested that VIRMA, as an m6A-related regulator, may significantly influence HNSCC progression by regulating ubiquitin protein ligase E3 component N-recognin 5 (UBR5) through m6A modification. Therefore, VIRMA may serve as a prognostic biomarker.

circRNA-PTPN4 mediated regulation of FOXO3 and ZO-1 expression: implications for blood-brain barrier integrity and cognitive function in uremic encephalopathy.

Uremic encephalopathy (UE) poses a significant challenge in neurology, leading to the need to investigate the involvement of non-coding RNA (ncRNA) in its development. This study employed ncRNA-seq and RNA-seq approaches to identify fundamental ncRNAs, specifically circRNA and miRNA, in the pathogenesis of UE using a mouse model. In vitro and in vivo experiments were conducted to explore the circRNA-PTPN4/miR-301a-3p/FOXO3 axis and its effects on blood-brain barrier (BBB) function and cognitive abilities. The research revealed that circRNA-PTPN4 binds to and inhibits miR-301a-3p, leading to an increase in FOXO3 expression. This upregulation results in alterations in the transcriptional regulation of ZO-1, affecting the permeability of human brain microvascular endothelial cells (HBMECs). The axis also influences the growth, proliferation, and migration of HBMECs. Mice with UE exhibited cognitive deficits, which were reversed by overexpression of circRNA-PTPN4, whereas silencing FOXO3 exacerbated these deficits. Furthermore, the uremic mice showed neuronal loss, inflammation, and dysfunction in the BBB, with the expression of circRNA-PTPN4 demonstrating therapeutic effects. In conclusion, circRNA-PTPN4 plays a role in promoting FOXO3 expression by sequestering miR-301a-3p, ultimately leading to the upregulation of ZO-1 expression and restoration of BBB function in mice with UE. This process contributes to the restoration of cognitive abilities.

Toward Accurate Human Parsing Through Edge Guided Diffusion.

Existing human parsing frameworks commonly employ joint learning of semantic edge detection and human parsing to facilitate the localization around boundary regions. Nevertheless, the parsing prediction within the interior of the part contour may still exhibit inconsistencies due to the inherent ambiguity of fine-grained semantics. In contrast, binary edge detection does not suffer from such fine-grained semantic ambiguity, leading to a typical failure case where misclassification occurs inner the part contour while the semantic edge is accurately detected. To address these challenges, we develop a novel diffusion scheme that incorporates guidance from the detected semantic edge to mitigate this problem by propagating corrected classified semantics into the misclassified regions. Building upon this diffusion scheme, we present an Edge Guided Diffusion Network (EGDNet) for human parsing, which can progressively refine the parsing predictions to enhance the accuracy and coherence of human parsing results. Moreover, we design a horizontal-vertical aggregation to exploit inherent correlations among body parts along both the horizontal and vertical axes, which aims at enhancing the initial parsing results. Extensive experimental evaluations on various challenging datasets demonstrate the effectiveness of the proposed EGDNet. Remarkably, our EGDNet shows impressive performances on six benchmark datasets, including four human body parsing datasets (LIP, CIHP, ATR, and PASCAL-Person-Part), and two human face parsing datasets (CelebAMask-HQ and LaPa).

pH-Responsive Pesticide-Loaded Hollow Mesoporous Silica Nanoparticles with ZnO Quantum Dots as a Gatekeeper for Control of Rice Blast Disease.

Nanotechnology-enabled pesticide delivery systems have been widely studied and show great prospects in modern agriculture. Nanodelivery systems not only achieve the controlled release of agrochemicals but also possess many unique characteristics. This study presents the development of a pH-responsive pesticide nanoformulation utilizing hollow mesoporous silica nanoparticles (HMSNs) as a nanocarrier. The nanocarrier was loaded with the photosensitive pesticide prochloraz (Pro) and then combined with ZnO quantum dots (ZnO QDs) through electrostatic interactions. ZnO QDs serve as both the pH-responsive gatekeeper and the enhancer of the pesticide. The results demonstrate that the prepared nanopesticide exhibits high loading efficiency (24.96%) for Pro. Compared with Pro technical, the degradation rate of Pro loaded in HMSNs@Pro@ZnO QDs was reduced by 26.4% after 24 h ultraviolet (UV) exposure, indicating clearly improved photostability. In a weak acidic environment (pH 5.0), the accumulated release of the nanopesticide after 48 h was 2.67-fold higher than that in a neutral environment. This indicates the excellent pH-responsive characteristic of the nanopesticide. The tracking experiments revealed that HMSNs can be absorbed by rice leaves and subsequently transported to other tissues, indicating their potential for effective systemic distribution and targeted delivery. Furthermore, the bioactivity assays confirmed the fungicidal efficacy of the nanopesticide against rice blast disease. Therefore, the constructed nanopesticide holds great prospect in nanoenabled agriculture, offering a novel strategy to enhance pesticide utilization.

Comprehensive Attribute Prediction Learning for Person Search by Language.

Person search by language refers to searching for the interested pedestrian images given natural language sentences, which requires capturing fine-grained differences to accurately distinguish different pedestrians, while still far from being well addressed by most of the current solutions. In this paper, we propose the Comprehensive Attribute Prediction Learning (CAPL) method, which explicitly carries out attribute prediction learning, for improving the modeling capabilities of fine-grained semantic attributes and obtaining more discriminative visual and textual representations. First, we construct the semantic ATTribute Vocabulary (ATT-Vocab) based on sentence analysis. Second, the complementary context-wise and attribute-wise attribute predictions are simultaneously conducted to better model the high-frequency in-vocab attributes in our In-vocab Attribute Prediction (IAP) module. Third, to additionally consider the out-of-vocab semantics, we present the Attribute Completeness Learning (ACL) module for better capturing the low-frequency attributes outside the ATT-Vocab, obtaining more comprehensive representations. Combining the IAP and ACL modules together, our CAPL method has obtained the currently state-of-the-art retrieval performance on two widely-used benchmarks, i.e., CUHK-PEDES and ICFG-PEDES datasets. Extensive experiments and analyses have been carried out to validate the effectiveness and generalization capacities of our CAPL method.

Context Recovery and Knowledge Retrieval: A Novel Two-Stream Framework for Video Anomaly Detection.

Video anomaly detection aims to find the events in a video that do not conform to the expected behavior. The prevalent methods mainly detect anomalies by snippet reconstruction or future frame prediction error. However, the error is highly dependent on the local context of the current snippet and lacks the understanding of normality. To address this issue, we propose to detect anomalous events not only by the local context, but also according to the consistency between the testing event and the knowledge about normality from the training data. Concretely, we propose a novel two-stream framework based on context recovery and knowledge retrieval, where the two streams can complement each other. For the context recovery stream, we propose a spatiotemporal U-Net which can fully utilize the motion information to predict the future frame. Furthermore, we propose a maximum local error mechanism to alleviate the problem of large recovery errors caused by complex foreground objects. For the knowledge retrieval stream, we propose an improved learnable locality-sensitive hashing, which optimizes hash functions via a Siamese network and a mutual difference loss. The knowledge about normality is encoded and stored in hash tables, and the distance between the testing event and the knowledge representation is used to reveal the probability of anomaly. Finally, we fuse the anomaly scores from the two streams to detect anomalies. Extensive experiments demonstrate the effectiveness and complementarity of the two streams, whereby the proposed two-stream framework achieves state-of-the-art performance on ShanghaiTech, Avenue and Corridor datasets among the methods without object detection. Even if compared with the methods using object detection, our method reaches competitive or better performance on the ShanghaiTech, Avenue, and Ped2 datasets.

Establishment of transgenic fluorescent mice for labeling synapses and screening synaptogenic adhesion molecules.

Synapse is the fundamental structure for neurons to transmit information between cells. The proper synapse formation is crucial for developing neural circuits and cognitive functions of the brain. The aberrant synapse formation has been proved to cause many neurological disorders, including autism spectrum disorders and intellectual disability. Synaptic cell adhesion molecules (CAMs) are thought to play a major role in achieving mechanistic cell-cell recognition and initiating synapse formation via trans-synaptic interactions. Due to the diversity of synapses in different brain areas, circuits and neurons, although many synaptic CAMs, such as Neurexins (NRXNs), Neuroligins (NLGNs), Synaptic cell adhesion molecules (SynCAMs), Leucine-rich-repeat transmembrane neuronal proteins (LRRTMs), and SLIT and NTRK-like protein (SLITRKs) have been identified as synaptogenic molecules, how these molecules determine specific synapse formation and whether other molecules driving synapse formation remain undiscovered are unclear. Here, to provide a tool for synapse labeling and synaptic CAMs screening by artificial synapse formation (ASF) assay, we generated synaptotagmin-1-tdTomato (Syt1-tdTomato) transgenic mice by inserting the tdTomato-fused synaptotagmin-1 coding sequence into the genome of C57BL/6J mice. In the brain of Syt1-tdTomato transgenic mice, the tdTomato-fused synaptotagmin-1 (SYT1-tdTomato) signals were widely observed in different areas and overlapped with synapsin-1, a widely-used synaptic marker. In the olfactory bulb, the SYT1-tdTomato signals are highly enriched in the glomerulus. In the cultured hippocampal neurons, the SYT1-tdTomato signals showed colocalization with several synaptic markers. Compared to the wild-type (WT) mouse neurons, cultured hippocampal neurons from Syt1-tdTomato transgenic mice presented normal synaptic neurotransmission. In ASF assays, neurons from Syt1-tdTomato transgenic mice could form synaptic connections with HEK293T cells expressing NLGN2, LRRTM2, and SLITRK2 without immunostaining. Therefore, our work suggested that the Syt1-tdTomato transgenic mice with the ability to label synapses by tdTomato, and it will be a convenient tool for screening synaptogenic molecules.

Toward Video Anomaly Retrieval From Video Anomaly Detection: New Benchmarks and Model.

Video anomaly detection (VAD) has been paid increasing attention due to its potential applications, its current dominant tasks focus on online detecting anomalies, which can be roughly interpreted as the binary or multiple event classification. However, such a setup that builds relationships between complicated anomalous events and single labels, e.g., "vandalism", is superficial, since single labels are deficient to characterize anomalous events. In reality, users tend to search a specific video rather than a series of approximate videos. Therefore, retrieving anomalous events using detailed descriptions is practical and positive but few researches focus on this. In this context, we propose a novel task called Video Anomaly Retrieval (VAR), which aims to pragmatically retrieve relevant anomalous videos by cross-modalities, e.g., language descriptions and synchronous audios. Unlike the current video retrieval where videos are assumed to be temporally well-trimmed with short duration, VAR is devised to retrieve long untrimmed videos which may be partially relevant to the given query. To achieve this, we present two large-scale VAR benchmarks and design a model called Anomaly-Led Alignment Network (ALAN) for VAR. In ALAN, we propose an anomaly-led sampling to focus on key segments in long untrimmed videos. Then, we introduce an efficient pretext task to enhance semantic associations between video-text fine-grained representations. Besides, we leverage two complementary alignments to further match cross-modal contents. Experimental results on two benchmarks reveal the challenges of VAR task and also demonstrate the advantages of our tailored method. Captions are publicly released at https://github.com/Roc-Ng/VAR.

Unsupervised Test-Time Adaptation Learning for Effective Hyperspectral Image Super-Resolution With Unknown Degeneration.

Fusing a low-resolution hyperspectral image (HSI) with a high-resolution (HR) multi-spectral image has provided an effective way for HSI super-resolution (SR). The key lies on inferring the posteriori of the latent (i.e., HR) HSI using an appropriate image prior and the likelihood determined by the degeneration between the latent HSI and the observed images. However, in scenarios with complex imaging environments and various imaging scenes, the prior of HSIs can be prohibitively complicated and the degeneration is often unknown, which causes it difficult to accurately infer the posteriori of each latent HSI. To tackle this problem, we present an unsupervised test-time adaptation learning (UTAL) framework for HSI SR under unknown degeneration. Instead of directly modeling the complicated image prior, it first implicitly learns a content-agnostic prior shared across different images through supervisedly pre-training a mutual-guiding fusion module on extensive synthetic data. Then, it adapts the shared prior to those private characteristics in the latent HSI for posteriori inference through unsupervisedly learning a self-guiding adaptation module and a degeneration estimation network on two observed images in the test phase. Such a two-stage learning scheme models the complicated image prior in a divide-and-conquer manner, which eases the modeling difficulty and improves the prior accuracy. Moreover, the unknown degeneration can be estimated properly. Both of these two advantages empower us to accurately infer the posteriori of the latent HSI, thereby increasing the generalization performance in real applications. Additionally, in order to further mitigate the over-fitting in coping with more challenging cases (e.g., degenerations in both spectral and spatial domains are unknown) and speed up, we propose to meta-train UTAL on extensive synthetic SR tasks and solve it using an alternative optimization strategy such that UTAL learns to produce good generalization performance in real challenging cases with a small number of gradient descent steps. To verify the efficacy of UTAL, we evaluate it on HSI SR tasks with different unknown degenerations as well as some other HSI restoration tasks (e.g., compressive sensing), and report strong results superior to that of existing competitors.

The development of local ambient air quality standards: A case study of Hainan Province, China.

The ambient air quality standard (AAQS) is a vital policy instrument for protecting the environment and human health. Hainan Province is at the forefront of China's efforts to protect its ecological environment, with an official goal to achieve world-leading air quality by 2035. However, neither the national AAQS nor the World Health Organization guideline offers sufficient guidance for improving air quality in Hainan because Hainan has well met the former while the latter is excessively stringent. Consequently, the establishment of Hainan's local AAQS becomes imperative. Nonetheless, research regarding the development of local AAQS is scarce, especially in comparatively more polluted countries such as China. The relatively high background values and significant interannual fluctuations in air pollutant concentrations in Hainan present challenges in the development of local AAQS. Our research proposes a world-class local AAQS of Hainan Province by reviewing the AAQS in major countries or regions worldwide, analyzing the influence of different statistical forms, and carefully evaluating the attainability of the standard. In the proposed AAQS, the annual mean concentration limit for PM2.5, the annual 95th percentile of daily maximum 8-h mean (MDA8) concentration limit for O3, and the peak season concentration limit for O3 are set at 10, 120, and 85 µg/m3, respectively. Our study indicates that, with effective control policies, Hainan is projected to achieve compliance with the new standard by 2035. The implementation of the local AAQS is estimated to avoid 1,526 (1,253-1,789) and 259 (132-501) premature deaths attributable to long-term exposure to PM2.5 and O3 in Hainan in 2035, respectively.

Response of organic aerosol in Beijing to emission reductions during the XXIV Olympic Winter Games.

Organic aerosol (OA) serves as a crucial component of fine particulate matter. However, the response of OA to changes in anthropogenic emissions remains unclear due to its complexity. The XXIV Olympic Winter Games (OWG) provided real atmospheric experimental conditions on studying the response of OA to substantial emission reductions in winter. Here, we explored the sources and variations of OA based on the observation of aerosol mass spectrometer (AMS) combined with positive matrix factorization (PMF) analysis in urban Beijing during the 2022 Olympic Winter Games. The influences of meteorological conditions on OA concentrations were corrected by CO and verified by deweathered model. The CO-normalized primary OA (POA) concentrations from traffic, cooking, coal and biomass burning during the OWG decreased by 39.8 %, 23.2 % and 65.0 %, respectively. Measures controlling coal and biomass burning were most effective in reducing POA during the OWG. For the CO-normalized concentration of secondary OA (SOA), aqueous-phase related oxygenated OA decreased by 51.8 % due to the lower relative humidity and emission reduction in precursors, while the less oxidized­oxygenated OA even slightly increased as the enhanced atmospheric oxidation processes may partially offset the efficacy of emission control. Therefore, more targeted reduction of organic precursors shall be enhanced to lower atmospheric oxidation capacity and mitigate SOA pollution.

Identification of Genes Associated with Decreasing Abundance of Monocytes in Long-Term Peritoneal Dialysis Patients.

Purpose: Chronic kidney disease (CKD) will become an end-stage renal disease (ESRD) at stage 5. Peritoneal dialysis (PD) is required for renal replacement therapy. This study aims to identify monocytes-related genes in peritoneal cells from long-term PD (LPD) patients and short-term PD (SPD) patients. Methods: Bulk RNA-seq data (GSE125498 dataset) and ScRNA-seq data (GSE130888) were downloaded to identify differentially expressed genes, monocytes-related genes, and monocytes marker genes in LPD patients. Immune infiltration was analyzed in the GSE125498 dataset. Core genes associated with monocytes changes were screened out, followed by functional analysis and expression validation using RT-PCR. Results: Monocytes are the most abundant immune cell in PD. The number of monocytes was remarkably decreased in LPD compared with SPD. A total of 16 up-regulated core genes negatively correlated with the abundance of monocytes were obtained in LPD. The expression of 16 core genes was lower in monocyte clusters than that in other cell clusters. In addition, LCK, CD3G, CD3E, CD3D, and LAT were involved in the signaling pathways of Th1 and Th2 cell differentiation, T cell receptor signaling pathway, and Th17 cell differentiation. CD2 was involved in hematopoietic cell lineage signaling pathway. Conclusion: Identification of monocytes related-genes and related signaling pathways could be helpful in understanding the molecular mechanism of monocytes changes during PD.

Nitrided Rhodium Nanoclusters with Optimized Water Bonding and Splitting Effects for pH-Universal H2 -Production.

The nitridation of noble metals-based catalysts to further enhance their hydrogen evolution reaction (HER) kinetics in neutral and alkaline conditions would be an effective strategy for developing high-performance wide pH HER catalysts. Herein, a facile molten urea method is employed to construct the nitrided Rh nanoclusters (Rhx N) supported on N-doped carbon (Rhx N-NC). The uniformly distributed Rhx N clusters exhibited optimized water bonding and splitting effects, therefore resulting in excellent pH-universal HER performance. The optimized Rhx N-NC catalyst only requires 8, 12, and 109 mV overpotentials to reach the current density of 10 mA cm-2 in 0.5 M H2 SO4 , 1.0 M KOH, and 1.0 M PBS electrolytes, respectively. The spectroscopic characterizations and theoretical calculation further confirm the vital role of Rh-N moieties in Rhx N clusters in improving the transfer of electrons and facilitating the generation of H2 . This work not only provides a suitable nitridation method for noble metal species in mild conditions but also makes a breakthrough in synthesizing noble metal nitrides-based electrocatalysts to achieve an exceptional wide-pH HER performance and other catalysis.

PSNet: A Deep Learning Model-Based Single-Shot Digital Phase-Shifting Algorithm.

In contrast to traditional phase-shifting (PS) algorithms, which rely on capturing multiple fringe patterns with different phase shifts, digital PS algorithms provide a competitive alternative to relative phase retrieval, which achieves improved efficiency since only one pattern is required for multiple PS pattern generation. Recent deep learning-based algorithms further enhance the retrieved phase quality of complex surfaces with discontinuity, achieving state-of-the-art performance. However, since much attention has been paid to understanding image intensity mapping, such as supervision via fringe intensity loss, global temporal dependency between patterns is often ignored, which leaves room for further improvement. In this paper, we propose a deep learning model-based digital PS algorithm, termed PSNet. A loss combining both local and global temporal information among the generated fringe patterns has been constructed, which forces the model to learn inter-frame dependency between adjacent patterns, and hence leads to the improved accuracy of PS pattern generation and the associated phase retrieval. Both simulation and real-world experimental results have demonstrated the efficacy and improvement of the proposed algorithm against the state of the art.

Corrigendum: Interferon-induced transmembrane protein 3 in the hippocampus: a potential novel target for the therapeutic effects of recombinant human brain natriuretic peptide on sepsis-associated encephalopathy.

[This corrects the article DOI: 10.3389/fnmol.2023.1182005.].

Interferon-induced transmembrane protein 3 in the hippocampus: a potential novel target for the therapeutic effects of recombinant human brain natriuretic peptide on sepsis-associated encephalopathy.

Objective: This study aims to explore whether interferon-induced transmembrane protein 3 (IFITM3) is involved in recombinant human brain natriuretic peptide (rhBNP)-mediated effects on sepsis-induced cognitive dysfunction in mice. Methods: The cellular localization and expression level of IFITM3 in the hippocampus were detected. The IFITM3 overexpression was achieved using an intracranial stereotactic system to inject an adeno-associated virus into the hippocampal CA1 region of mice. Field experiments, an elevated plus maze, and conditioned fear memory tests assessed the cognitive impairment in rhBNP-treated septic mice. Finally, in the hippocampus of septic mice, terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) staining and Immunoblot were used to detect changes in the protein expression of cleaved Caspase-8 and cleaved Caspase-3 in apoptosis-related pathways, and toll-like receptor 4 (TLR4) and nuclear factor κB (NF-κB) p65 in inflammatory pathways. Results: Fourteen days after cecal ligation and puncture (CLP) surgery, IFITM3 localized in the plasma membrane and cytoplasm of the astrocytes in the hippocampus of septic mice, partially attached to the perivascular and neuronal surfaces, but not expressed in the microglia. The expression of IFITM3 was increased in the astrocytes and neurons in the hippocampus of septic mice, which was selectively inhibited by the administration of rhBNP. Overexpression of IFITM3 resulted in elevated anxiety levels and long-term learning and memory dysfunction, completely abolished the therapeutic effect of rhBNP on cognitive impairment in septic mice, and induced an increase in the number of neuronal apoptosis in the hippocampal CA1 region. The expression levels of cleaved Caspase-3 and cleaved Caspase-8 proteins were significantly increased in the hippocampus, but the expression levels of TLR4 and NF-κB p65 were not increased. Conclusion: The activation of IFITM3 may be a potential new target for treating sepsis-associated encephalopathy (SAE), and it may be one of the key anti-apoptotic mechanisms in rhBNP exerting its therapeutic effect, providing new insight into the clinical treatment of SAE patients.

Task-Incremental Learning for Drone Pilot Identification Scheme.

With the maturity of Unmanned Aerial Vehicle (UAV) technology and the development of Industrial Internet of Things, drones have become an indispensable part of intelligent transportation systems. Due to the absence of an effective identification scheme, most commercial drones suffer from impersonation attacks during their flight procedure. Some pioneering works have already attempted to validate the pilot's legal status at the beginning and during the flight time. However, the off-the-shelf pilot identification scheme can not adapt to the dynamic pilot membership management due to a lack of extensibility. To address this challenge, we propose an incremental learning-based drone pilot identification scheme to protect drones from impersonation attacks. By utilizing the pilot temporal operational behavioral traits, the proposed identification scheme could validate pilot legal status and dynamically adapt newly registered pilots into a well-constructed identification scheme for dynamic pilot membership management. After systemic experiments, the proposed scheme was capable of achieving the best average identification accuracy with 95.71% on P450 and 94.23% on S500. With the number of registered pilots being increased, the proposed scheme still maintains high identification performance for the newly added and the previously registered pilots. Owing to the minimal system overhead, this identification scheme demonstrates high potential to protect drones from impersonation attacks.

Curing Mechanisms of Polymeric Nano-Copolymer Subgrade.

The mechanical properties of the subgrade have a significant impact on the service life and pavement performance of the superstructure of pavement. By adding admixtures and via other means to strengthen the adhesion between soil particles, the strength and stiffness of the soil can be improved to ensure the long-term stability of pavement structures. In this study, a mixture of polymer particles and nanomaterials was used as a curing agent to examine the curing mechanism and mechanical properties of subgrade soil. Using microscopic experiments, the strengthening mechanism of solidified soil was analyzed with scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), Fourier infrared spectroscopy (FTIR), and X-ray diffraction (XDR). The results showed that with the addition of the curing agent, small cementing substances on the surface of soil minerals filled the pores between minerals. At the same time, with an increase in the curing age, the colloidal particles in the soil increased, and some of them formed large aggregate structures that gradually covered the surface of the soil particles and minerals. By enhancing the cohesiveness and integrity between different particles, the overall structure of the soil became denser. Through pH tests, it was found that the age had a certain effect on the pH of solidified soil, but the effect was not obvious. Through the comparative analysis of elements in plain soil and solidified soil, it was found that no new chemical elements were produced in the solidified soil, indicating that the curing agent does not have negative impacts on the environment.

Improving Inconspicuous Attributes Modeling for Person Search by Language.

Person search by language aims to retrieve the interested pedestrian images based on natural language sentences. Although great efforts have been made to address the cross-modal heterogeneity, most of the current solutions suffer from only capturing salient attributes while ignoring inconspicuous ones, being weak in distinguishing very similar pedestrians. In this work, we propose the Adaptive Salient Attribute Mask Network (ASAMN) to adaptively mask the salient attributes for cross-modal alignments, and therefore induce the model to simultaneously focus on inconspicuous attributes. Specifically, we consider the uni-modal and cross-modal relations for masking salient attributes in the Uni-modal Salient Attribute Mask (USAM) and Cross-modal Salient Attribute Mask (CSAM) modules, respectively. Then the Attribute Modeling Balance (AMB) module is presented to randomly select a proportion of masked features for cross-modal alignments, ensuring the balance of modeling capacity of both salient attributes and inconspicuous ones. Extensive experiments and analyses have been carried out to validate the effectiveness and generalization capacity of our proposed ASAMN method, and we have obtained the state-of-the-art retrieval performance on the widely-used CUHK-PEDES and ICFG-PEDES benchmarks.